Current Status of Corporate Income Tax and FDI Policies in Vietnam

foreign branches for tax avoidance purposes, coordinated sharing of tax information, management of permanent establishment formation… Implementing these actions requires coordination at various levels between countries and territories globally to effectively deal with BEPS practices.

4.2 Current status of corporate income tax and FDI policies in Vietnam

Vietnam is one of the developing countries that has achieved many positive results in attracting FDI in the period 2009-2019, when the annual FDI inflow increased by nearly 1000% over the past 10 years. FDI has truly become the main driving force for economic development and contributed to Vietnam's impressive economic growth rate, averaging 6.5-7%/year over the past decade. FDI has created resources for the government and jobs for the people. With a population of nearly 100 million people, Vietnam has become a very attractive market for foreign investors. Moreover, the system of signed free trade agreements has also helped Vietnam improve its competitiveness in accessing other markets. These are factors that can help Vietnam become an investment attraction center in the region. During the period from 1988 to 2019, FDI capital attracted to Vietnam has averaged more than 7 billion USD/year, an average of about 2.2 million USD/person. Of which, the period 2011-2019 attracted the most projects and total registered FDI capital (Table 4.2).

Table 4.2 Results of FDI attraction in Vietnam over time

Stage

Project Number	Total registered capital (million USD)	Total implemented capital (million USD)
Total	33,921	454,019.0	211,472.9
1988-1999	3.164	42,729.3	18,269.5
2000-2010	10,473	171,643.3	60,876.7
2011-2019	20,495	239,646.4	132,326.7

Maybe you are interested!

Source: General Statistics Office (2020)

However, in the period 1988-2019, FDI capital was about 47% of registered capital. Of which, FDI capital from countries with average technology and not holding high technology accounted for a large proportion, while capital from the world's leading developed countries was not really effective: FDI capital from the US was 2.7%, from Germany about 0.6%, from the UK about 1%, from France about 0.9%, and mainly from Japan and South Korea (35%).

Therefore, if we are not selective and only attract FDI from countries that do not possess high technology, Vietnam will not have high technology and will only attract FDI in the commercial sector and sectors with only medium technology.

Regarding state budget revenues, revenues from FDI enterprises are also increasing, from about 10.8% in 2010 to about 14.6% in 2020 (Table 4.3). This is considered a good trend, but it is still not commensurate with the actual potential of the FDI sector because there are still many limitations in tax management for FDI that need to be overcome. In the export sector, the FDI sector makes an important contribution, from about 54.2% in 2010 to about 70.6% in 2015 and over 71% in 2020. Thus, the FDI sector also makes a great contribution to increasing the openness of the economy.

Table 4.3 Contribution of FDI to Vietnam's state budget revenue and exports

Target

Unit	2010	2015	2019	Comparison of period 2010 -2019
1. State budget revenue, current price	trillion dong	599.9	1,020.5	1,551.1
- Revenue from FDI enterprises only	trillion dong	64.9	141	210.2	+ 148.3
% of total state budget revenue	%	10.8	13.9	13.6	+ 2.8%
2. Total export value	Billion USD	72.2	162.0	264.2
- Only for FDI sector	Billion USD	39.1	114.4	179.2	+ 140.1
% of total	%	54.2	70.6	67.8	+ 13.6%

Source: Author processed according to data from General Statistics Office

Regarding tax policies to attract FDI, Vietnam has made many adjustments and changes in the period 2009-2019. The most important change to increase tax competitiveness and attract investment is the policy of reducing the common corporate income tax rate according to the law. This tax rate was adjusted down to 25% from January 1, 2009 according to the Law on Corporate Income Tax (amended) in 2008. During this period, newly established enterprises continued to enjoy preferential corporate income tax rates of 10%, 15% and 20% depending on the investment sector or preferential investment location. In addition, enterprises also enjoy tax-free income incentives for some income from scientific research contracts, from the production of experimental products, etc. After the amendments and supplements in 2013 and 2014, the common corporate income tax rate was reduced from 25% to 22% (from

1/1/2014) and down to 20% (from 1/1/2016). The Law amending and supplementing a number of articles of the Law on Corporate Income Tax in 2013 has added incentives for investment in industrial parks (except for industrial parks in areas with favorable socio-economic conditions) and expansion investment projects. Next, the Law amending and supplementing a number of articles of tax laws No. 71/2014/QH13, effective from 01/01/2015, has added a number of fields and industries eligible for tax incentives such as: Cultivation, animal husbandry, processing of agricultural, forestry and aquatic products (incentives are not applied to the field of processing forest products); production of supporting industrial products; large-scale and high-tech production projects. The important change in investment incentives since 2013 is to change the investment incentives for newly established enterprises from investment projects in preferential sectors or locations to incentives for income from investment projects in preferential investment sectors or locations. In fact, although applying the same corporate income tax rate, compared to domestic enterprises, FDI enterprises still have some advantages in capital and modern production technology. Therefore, FDI enterprises still enjoy some tax incentives such as in the fields of high technology, environmental protection, agricultural, forestry and fishery processing, and supporting industries.

However, transfer pricing and tax evasion by foreign direct investment enterprises remain a difficult problem. To overcome this situation, the Government has issued Decree 20/2017/ND-CP on tax management for enterprises with related-party transactions, which is still being revised and supplemented, in order to both comply with international practices and to inspect, control and ensure foreign investors when investing in Vietnam.

Thus, in recent times, the preferential corporate income tax policy has had a positive impact on FDI enterprises. Accordingly, enterprises investing in Vietnam have both conditions to develop production and business and help Vietnam's economic growth. However, FDI in Vietnam still has many limitations and problems such as transfer pricing, tax evasion, environmental pollution; the spillover effects and linkages between the FDI sector and the domestic sector have not been as expected; the orientation of attracting FDI by industry and partner is still limited; although many good technologies have been attracted, the goal of attracting high technology, source technology and technology transfer has not been achieved...

4.3 Research results and discussion

4.3.1 Causal relationship between corporate income tax and FDI

The main objective of the thesis is to study the relationship between corporate income tax and FDI. This is a two-way relationship, which is expressed through the impact of corporate income tax on FDI, and the impact of FDI (in the role of tax havens) on the corporate income tax base. Therefore, the thesis will first evaluate the two-way Granger causality relationship between these two pairs of variables. The Granger causality relationship between corporate income tax and FDI is tested for the relationship between the variables:

(i) The relationship between corporate income tax revenue, corporate income tax base and FDI inflows to developing countries.

(ii) The relationship between corporate income tax base and FDI flows from tax havens to developing countries.

4.3.1.1 Testing for inter-country dependence

The test is first conducted to determine the existence of a relationship between countries in the region (Urbain & Westerlund, 2006). The thesis uses Lagarane multiplier statistics through Pesaran (2004), Friedman (1937) and Frees (1995) statistics used for the fixed effects model (Hoyos & Sarafidis, 2006; Blackburne & Frank, 2007) in the process of testing the dependence between countries.

Table 4.4: Results of testing for inter-country dependence

DEPENDENT VARIABLE

INDEPENDENT VARIABLE	CD Test (Pesaran, 2004)	Friedman test (1973)	Frees test (1995)
LFDI	TAXINCOME; RINCOME	36.517*** (Pr = 0.000)	152.091*** (Pr = 0.000)	8,067
RINCOME	LFDI; RHAVE	9.334*** (Pr = 0.000)	48.722** (Pr = 0.022)	3,933

*, **, *** correspond to statistical significance levels of 10%, 5%, 1%, respectively.

Source: Compiled and calculated from Stata The test results show that there is sufficient evidence to reject the hypothesis of independence between countries at the 1% significance level in all three proposed tests. This finding requires that unit root tests for stationary data must take into account the dependence between

countries in the panel data.

4.3.1.2 Stationarity test for panel data

Since Granger causality test requires stationary series, the variables involved will be tested for stationarity for panel data. The author checks the stationarity of the variables using Augmented Dickey Fuller (ADF), Phillips-Perron (PP) and Im- Pesaran-Shin (IPS) tests. The null hypothesis of this test is that there is a unit root, meaning the variables are not stationary.

Table 4.5: Stationarity test for panel data

Variable

ADF		PP		IPS
	Coefficient	Coefficients and trends	Coefficient	Coefficients and trends	Coefficient	Coefficients and trends
LFDI	-9.103 ***	-9.461 ***	-4.266 ***	-4.962 ***	-2.9385 ***	-5.791 ***
RHAVE	-11,063 ***	-12,026 ***	-4.137 ***	-4.811 ***	-2.6931 ***	-4.721 ***
RINCOME	-10,754 ***	-11,497 ***	-6.409 ***	-7.711 ***	-3.3547 ***	-6.052 ***
TAXINCOME	-7.963 ***	-8,070 ***	-2.811 ***	-3.614 ***	-2.8837 ***	-5.531 ***

*, **, *** correspond to statistical significance levels of 10%, 5%, 1%, respectively.

Source: Synthesized and calculated from Stata The stationarity test results are shown in Table 4.5, the stationarity tests of all variables are stationary at the original level for all non-trend tests and

tend

4.3.1.3 Wasterlund's cointegration test for panel data

Next, based on the Westerlund (2008) cointegration test, the thesis examines the long-run relationships between the variables because of the cross-sectional dependence that appears in the econometric analysis of the data set. The cointegration test will also help determine the appropriate lag through which the two variables have the best cointegration.

To choose the optimal lag for the Granger test, the author based on the proposal of Atukeren (2007) and Hartwwig (2009) used the AIC (Akaike's information Criterion) and SIC (Schwaz Information Criterion) criteria. According to these criteria, the variables have an optimal lag of 1.

Table 4.6: Westerlund cointegration test Dependent variable: LFDI (Lag =1)

Independent variable

Gt	Station	Pt	Pa
RINCOME	-4.395***	-19,796***	-15,576***	-16.874***
AXYNCOME	-4.324***	-24,577***	-20,970***	-25.065***

The results show that the hypothesis of no cointegration for the dependent variable: LFDI and the variable RINCOME is rejected at the 1% significance level, and the hypothesis of no cointegration for the dependent variable: LFDI and the variable TAXINCOME is also rejected at the 1% significance level. Thus, there is a long-run relationship between FDI and the two income tax base variables and the income tax revenue variable mentioned above in the research sample of 32 developing countries.

Table 4.7: Westerlund cointegration test Dependent variable: RINCOME (Lag =1)

Independent variable

Gt	Station	Pt	Pa
LFDI	-5.339 ***	-18,266 ***	-17,488 ***	-16.125 ***
RHAVE	-7,599 ***	-18,601 ***	-21,276 ***	-19,195 ***

*, **, *** correspond to statistical significance levels of 10%, 5%, 1%, respectively.

Source: Synthesized and calculated from Stata Table 4.7 shows that: The hypothesis of no cointegration for the dependent variable RINCOME and the variable LFDI is rejected at the 1% significance level, and the hypothesis of no cointegration for the dependent variable RINCOME and the variable RHAVE is also rejected at the 1% significance level. Thus, there is a long-run relationship between FDI and the two income tax base variables (FDI income), and the variable FDI from tax havens to developing countries.

development in the study sample.

4.3.1.4 Testing causal relationships

The causality test of Dumitrescu & Hurlin (2012) is presented in Table 4.8 showing a bidirectional relationship between the variables: tax revenue and FDI, tax base and FDI, FDI income (CIT base) and tax haven.

Table 4.8: Test of causal relationship

HYPOTHESIS

W-Stat.	Zbar Stat
TAXINCOME → LFDI	1.625	2.502** (p-value = 0.012)
LFDI → TAXINCOME	4,278	13.113*** (p-value = 0.000)
RINCOME → LFDI	3.173	8.694*** (p-value = 0.000)
LFDI → RINCOME	4,550	14.203*** (p-value = 0.000)
RHAVE → RINCOME	2,260	5.041*** (p-value = 0.000)
RINCOME → RHAVE	3.191	8.766*** (p-value = 0.000)

Specific causal relationship test results:

For corporate income tax revenue (TAXINCOME): there is a bidirectional relationship between corporate income tax revenue and FDI. This result supports the studies of Aslam (2015) and Bayar & Ozturk (2018), which show a bidirectional causal relationship between foreign direct investment inflows and total corporate income tax revenue.

For the corporate income tax base (RINCOME): the tax base for calculating corporate income tax is the income of FDI enterprises, the results show a causal relationship between foreign direct investment inflows and FDI income. This is consistent with the theories of FDI and the practice of tax exemption agreements related to increased foreign direct investment (Azémar & Dharmapala, 2019). On the contrary, in tax avoidance schemes, MNCs have sought to reduce FDI income in developing countries, directly eroding the income tax base (BEPS).

The role of FDI flows from tax havens to developing countries: the research results show that there is a two-way relationship between FDI from tax havens (RHAVE) and the corporate income tax base (RINCOME). This result is completely similar to Crivelli E, De Mooij R, Keen M., (2016) and Richard Bolwijn, Bruno Casella and Davide Rig (2018) emphasizing that the shift of FDI capital through offshore financial centers (OFCs), especially tax havens, has caused the erosion of the BEPS tax base.

Thus, the results of the causal relationship between tax and FDI variables as above will be the basis for the thesis to continue to conduct further empirical studies on the relationship between tax and FDI.

4.3.2 Impact of corporate income tax policy on FDI

4.3.2.1 Estimated results

The estimated results in the research sample are shown in three columns corresponding to the models of tax impact on FDI including: statutory tax rate (column 1), effective tax rate (column 2), and corporate income tax revenue (column 3) in Table 4.9. When considering the suitability of the estimated model, in the process of analyzing the variables, it is shown that there will be endogeneity of the independent variable and the dependent variable (LFDI), or endogeneity between independent variables, so the author used the GMM method to reduce this problem. In the GMM method, the instrument variable used is often related to population or labor (Sachs & Warner, 1997; Frankel & Romer, 1999) in the analysis of factors affecting FDI.

simultaneously affect FDI. With the above characteristics of the model, it can be affirmed the suitability of the estimation method used (dif-GMM).

Table 4.9 : Regression results of the impact of corporate income tax policy on FDI

VARIABLE

MODEL STATUTORY TAX RATE (1)	MODEL EFFECTIVE TAX RATE (2)	MODEL CORPORATE INCOME TAX REVENUE (3)
RTAXINC	-3.927***
TAXPRO		-0.038***
TAXINCOME			0.028***
DLGFDI	0.786***	0.785***	0.339***
GDP	-0.002*	-0.012***	-0.012***
LABOR	2,507***	2.902***	1,061**
POPULA	-4.064***	-0.690	-2.314**
INFLAT	-0.0025***	-0.005***	-.0045***
PROPRI	0.002***	0.001*	0.003***
GOINTEG	-0.006***	-0.001*	-0.010***
TRADE	-0.006***	-0.001**	-0.014***
Number of observations	256	256	256
Number of instrumental variables	30	30	28
Country Number	32	32	32
AR Testing (2)	0.560	0.183	0.380
Sargan test	0.209	0.487	0.638
Hansen test	0.703	0.725	0.556

Note: ***, ** and * denote significance levels of 1%, 5% and 10% respectively.

Source: Synthesized and calculated from Stata

The results show that statutory tax rates and effective tax rates have opposite effects on FDI. This is consistent with the expectation of the first hypothesis of the thesis, regarding the corporate income tax competition policy in the process of attracting FDI, countries always reduce corporate income tax rates, because this is the policy that investors are interested in and easily attracts FDI. This result is also completely similar to the research of Devereux & ctg (2002); Demekas (2007), Sato (2012), when pointing out that in all situations using statutory tax rates or effective tax rates, there is a significant negative impact on foreign direct investment. Compared with the results of Tomonori Sato (2012) that a one percentage point reduction in statutory tax rates increases foreign direct investment in the host country by about 2.4%, the results of the thesis have a higher coefficient, which is 3.92%.

However, when comparing the correlation of the impact of two types of statutory tax rates and effective tax rates, the results of the thesis show that the statutory tax rate has a larger impact coefficient, proving that the elasticity with FDI is larger than the effective tax rate. This is also the opinion in the studies of Nicod`eme (2001), Bénassy-Quéré & ctg (2003),

Comment